In recent years, along with the development of portable electronic equipment such as mobile phones and notebook computers, and the commercialization of electric automobiles, there is an increasing demand for a miniaturized and lightweight secondary battery with a high capacity. At present, as a secondary battery with a high capacity satisfying this demand, a non-aqueous secondary battery such as a lithium secondary battery using LiCoO2 for a positive electrode and a carbon material for a negative electrode is being commercialized. Such a lithium secondary battery has a high energy density, and can be miniaturized and reduced in weight, so that it has been paid attention to as a power source of portable electronic equipment.
LiCoO2 used as a material for a positive electrode of the lithium secondary battery is easy to produce and handle, so that it is often used as a preferable active material. However, LiCoO2 is produced using cobalt, which is rare metal, as a material. Therefore, it is conceivable that a material shortage will become serious in the future. Furthermore, the price of cobalt itself is high and fluctuates greatly, so that it is desired to develop a material for a positive electrode that can be supplied stably at a low cost.
In view of the above, materials of a lithium-manganese oxide type are expected to be a prospective substitute for LiCoO2 as a material for a positive electrode of a lithium secondary battery. Among them, lithium-manganese oxides with a Spinel structure, such as Li2Mn4O9, Li4Mn5O12, and LiMn2O4, are being investigated. In particular, LiMn2O4 can be charged/discharged in a potential range in the vicinity of 4 V against Li metal. The use of LiMn2O4 is disclosed in at least the following (JP 6 (1994)-76824 A, JP 7 (1995)-73883 A, JP 7 (1995)-230802 A, JP 7 (1995)-245106 A, etc.).
The theoretical discharge capacity of LiCoO2 is 274 mAh/g. However, when deep charging/discharging is conducted, LiCoO2 is changed in phase to influence a cycle life. Therefore, in an actual lithium secondary battery, the practical discharge capacity falls in a range of 125 to 140 mAh/g.
In contrast, the theoretical discharge capacity of LiMn2O4 is 148 mAh/g. However, LiMn2O4 also is changed in phase during charging/discharging in the same way as in LiCoO2. Furthermore, in the case of using a carbon material as a negative active material, since the irreversible capacity of the carbon material is large, the discharge capacity that can be used in the case where LiMn2O4 is used actually for a battery is decreased to about 90 to 105 mAh/g. As is apparent from this, when LiMn2O4 is used as a positive active material, the battery capacity cannot be increased compared with the case where LiCoO2 is used as a positive active material.
Furthermore, the true density of LiCoO2 is 4.9 to 5.1 g/cm3, whereas the true density of LiMn2O4 is very low (i.e., 4.0 to 4.2 g/cm3). Therefore, considering the filing property as a positive active material, LiMn2O4 is more disadvantageous in terms of the capacity.
Furthermore, in a lithium secondary battery using LiMn2O4 as a positive active material, the structure of LiMn2O4 itself is unstable during charging/discharging. Therefore, there is a problem that the cycle characteristics of the LiMn2O4 type battery are worse than those of the LiCoO2 type battery.
In order to solve the above-mentioned problem, it also is considered that a layered lithium-manganese oxide of LiMnO2 or the like having a structure different from that of LiMn2O4 is used as a material for a positive electrode. However, as a result of the detailed study of this oxide by the inventors of the present invention, it was found that the properties such as the structure and characteristics are changed remarkably due to the composition of a compound, in particular, the presence of elements constituting the oxide other than Li and Mn, the kind thereof, and the ratio of quantity thereof, and the process in which the oxide is formed.
For example, in the case where the average valence of Mn approaches 3 due to the fluctuation of the composition of Spinel type lithium-manganese oxide (LiMn2O4), the crystal structure of the above-mentioned oxide is strained to cause a phase change from the Spinel structure of a cubic to a tetragonal, whereby LiMnO2 is formed. The phase change from the cubic to the tetragonal occurs along with charging/discharging in a potential range in the vicinity of 3 V with respect to lithium. Therefore, the lithium secondary battery using the Spinel type lithium-manganese oxide (LiMn2O4) as a material for a positive electrode cannot be used in the same way as in the above-mentioned lithium secondary battery that is charged/discharged at a voltage in the vicinity of 4 V.
Furthermore, in the case where the structure molar ratio (Li/Mn) is 1, due to the Jahn-Teller effect of trivalent Mn, the crystal structure of LiMnO2 exhibits an orthorhombic system.
This compound (LiMnO2) can be charged/discharged electrochemically at a Li quantity ratio of 0 to 1.0, which results in a discharge capacity of about 285 mAh/g in terms of theory. However, as the ratio of tetravalent Mn is increased during initial charging, a phase transition to a Spinel structure occurs. Therefore, the initial charge/discharge curve and the second and subsequent charge/discharge curves exhibit different shapes. In addition, the discharge capacity in the case where discharging is terminated at a voltage of 3.5 V or more is decreased remarkably from a theoretical value. Furthermore, the structure is changed with the movement of Mn during charging/discharging. Therefore, cycle durability is insufficient, and rapid charging/discharging cannot be conducted.
Therefore, in order to commercialize a layered lithium-manganese oxide such as LiMnO2, it is required to solve the problems involved in stabilization of a crystal structure, an increase in capacity due to the enhancement of reversibility of charging/discharging, and durability during a charge/discharge cycle.